ABSTRACT

Geostatistical mapping of soil properties in Ibiono, Itu and Ikono Local Government Areas of Akwa Ibom State was carried out for site- specific soil management. A digital elevation model (DEM) of the study area was acquired from United State Geological Survey (USGS) at 30m resolution. Three terrain attributes: slope gradient, plan curvature and elevation gradient were generated from the DEM. Elevation gradient map generated was used in the field to guide field sampling by cross-checking whether what is in the map is in the field (ground-truthing) with the aid of Global Positioning System (GPS). Modified conditioned Latin hypercube sampling technique was used in selecting observation points. Each observation point was purposefully selected to fall within the classes of generated elevation map. Soil samples were collected from each observation point at 0-30 cm depth using soil auger. A total of 133 soil samples were collected in the study area for laboratory analysis. The results of the distribution of soil characteristics showed that organic carbon was significantly lower (p <0.05) at lower elevation gradient (0-40 masl) than others (41-80 masl) and (81-121). Available (P) and exchangeable (Na) were significantly higher (p <0.05) in the higher elevation gradient (81-121 masl) than lower and middle elevation gradients. Exchangeable K of middle and higher elevation gradients was significantly higher (p <0.05) than lower elevation (0-40 masl). The results of simple linear regression analysis between the target variables and selected predictor variables (slope gradient, elevation gradient and plan curvature) showed that slope gradient explained 2 % variation of available P; elevation gradient explained 5.4 % of clay, 4.3 % of soil pH, 4.7 % of available P and 6.1 % of exchangeable (K). A combination of the three selected terrain attributes in multiple linear regression explained 6.4 % of clay, 4.3 % of pH, 12.4 % of available P and 11.4 % of exchangeable K. The results of semi-variogram analysis of measured values of organic carbon, total N and (ECEC) (non-residuals) showed that these variables exhibited significant spatial structure with exponential model as the best fitted model for total N and ECEC and spherical model model for organic carbon. The results of semi- variogram of residuals showed that, spherical model was the best fitted model for residual available (P), clay, pH and exchangeable K.  The results of spatial prediction maps showed that the predicted values of clay fraction were higher in Ibiono Ibom and lower in Ikono and Itu. The predicted values of soil pH varied from extremely acid to slightly acid in Ibiono Ibom and strongly acid to slightly acid in Itu and Ikono Local Government Areas. The predicted values of organic carbon were moderate in Ibiono and high in Itu and Ikono Local Government Areas. The predicted values of total N were very low in the Ibiono Ibom, low in Ikono and moderate in Itu Local Government Area. The predicted values of ECEC were low in all parts of the study area.